Audio processing systems exist that attempts to correct the pitch and tempo of a singer, modifying notes sung off key or out of tempo. Other existing audio processing systems automatically control the volume of a given audio signal so that it remains within a given range (not too high or low). Yet others modify speech signals for improved telephone comprehension by older adults. Those systems, however, do not attempt to eliminate artifacts in speech associated with speech impairments such as stuttering while speaking, lisps and vocal ticks that might occur involuntarily.
Techniques for automatic recognition of stutter in speech signals have been explored, but no automatic correction procedure has been described.
Honal and Schultz in “Automatic Disfluency Removal On Recognized Spontaneous Speech-Rapid Adaptation To Speaker-Dependent Disfluencies”, IEEE ICASSP 2005, describe a method for removing disfluent words and phrases from an utterance, but this method is applied after the speech signal has been transcribed into text, and does not handle speech impairment at the level of the speech signal.